• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.644-646
• /
• 2023

스마트팩토리는 협동 로봇과 같은 프로그래머블한 설비의 유기적인 협업을 통해 최적화된 공정을 수행한다. 따라서 수집되는 센서 데이터의 특징과 환경 조건의 복잡도가 높아, 예지보전을 위한 AI 소프트웨어의 개발 시 요구사항 기반의 체계적인 개발 및 검증이 필수적이다. 본 논문에서는 AI 소프트웨어의 요구사항을 사용자와 시스템 관점에서 정의하고, AI 모델 개발 프로세스와 스마트팩토리 예지보전 측면에서 분석한다. 도출된 요구사항을 CNN 기반의 협동 로봇 기어 마모 예측 모델의 개발에 적용하여 데이터 관리와 모델 신뢰성 관점의 요구사항을 분석 및 검증하였다.

• JOURNAL OF ELECTRICAL WORLD
• /
• no.7 s.127
• /
• pp.96-99
• /
• 1987

최근설비를 계획하는데 있어서는 설계에서 건설 운용, 갱신에 이르기까지 라이프사이클코스트, 미니멈의 관점에서 시스템토틀로서 검토되는 일이 많다. 그러한 의미에서 설비의 보전기술을 중요한 역할을 담당한다. 일반적으로 보전의 종류는 사후, 예방, 예지()의 각 보전에, 또 보전의 방식으로는 시간기준보전과 상태기준보전으로 대별할 수가 있다. 설비가 복잡하고 고도화되어가고 있는 가운데 경비효율, 가동율, 신뢰도 등 토틀 시스템 효율의 향상을 도모하기 위해 상태기준에 의한 예지보전이 요구되는 경향에 있다. 또 온라인으로 리어얼타임인 설비진단기술의 진전이 이를 가능케 해가고 있다. 여기서는 음행분석에 의한 설비진단에 대해 자가발전장치의 음향진단을 예로 하여 기술키로 한다.

• 대학교육
• /
• s.36
• /
• pp.122-126
• /
• 1988

• 건강소식
• /
• v.28 no.2 s.303
• /
• pp.42-43
• /
• 2004

• CDE review
• /
• v.22 no.1
• /
• pp.58-62
• /
• 2016

• The Monthly Diabetes
• /
• s.207
• /
• pp.37-41
• /
• 2007

• 한국지구물리탐사학회:학술대회논문집
• /
• 2005.05a
• /
• pp.93-101
• /
• 2005

Seismicity changes associated with foreshocks of large Korean historical earthquakes of MMI > VIII are investigated for earthquake prediction study. A number of tests showed that b-values of foreshocks associated with these large earthquakes are most stable for precursor period of 13 years before the earthquake and rectangular source area of $1.1^{\circ}$ by $1.1^{\circ}$ around the epicenter. The b-values of foreshocks for 11 large events of MMI > VIII for the above foreshock area and precursor period turns out to be smaller than the value of 0.36 for the whole historical earthquakes with average 0.27. Epicenters of these foreshocks of small b-values are distributed close to the location of the main large earthquake. These observations indicate a possibility of predicting large earthquakes by closely monitoring the change of b-value for an extended period over decades in the Korean peninsula.

• Journal of the Korean Geophysical Society
• /
• v.8 no.3
• /
• pp.115-121
• /
• 2005

Seismicity changes associated with foreshocks of large Korean historical earthquakes of MMI > VIII are investigated for earthquake prediction study. A number of tests showed that b-values of foreshocks associated with these large earthquakes are most stable for precursor period of 13 years before the earthquake and rectangular source area of 1.1o by 1.1o around the epicenter. The b-values of foreshocks for 11 large events of MMI > VIII for the above foreshock area and precursor period turns out to be smaller than the value of 0.36 for the whole historical earthquakes with average 0.27. Epicenters of these foreshocks of small b-values are distributed close to the location of the main large earthquake. These observations indicate a possibility of predicting large earthquakes by closely monitoring the change of b-value for an extended period over decades in the Korean peninsula.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.40 no.8
• /
• pp.721-724
• /
• 2016

Because of the inherent uncertainties caused by the manufacturing process variations, future loading conditions, and incomplete damage models, the lifetimes of mechanical structures under field conditions are significantly different from the results obtained in the laboratories. In this study, a dual sensor was developed to prognosticate the fatigue failure of structures under these uncertain conditions, and its effectiveness was demonstrated on a rectangular columnar structure under repeated uni-axial loading. The dual sensor is a slightly weaker structure embedded in the target structure, so that failure occurs in the sensor earlier than in the target structure. From the signal differences in the strain gauges in the embedded dual sensor, it is possible to differentiate between the normal status and warning status, even under variable loads.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.7
• /
• pp.799-804
• /
• 2011

A thermal cycling test was conducted on a chip resistor solder joint with real-time failure monitoring. In order to study the failure mechanism of the chip resistor solder joint during the test, the resistance between both ends of the resistor was monitored until the occurrence of failure. It was observed that the monitored resistance first fluctuated linearly according to the temperature change. The initial variation in the resistance occurred at the time during the cycle when there was a decrease in temperature. A more significant change in the resistance followed after a certain number of cycles, during the time when there was an increase in the temperature. In order to explain the failure patterns of the solder joint, a mechanism for the solder failure was suggested, and its validity was proved through FE simulations. Based on the explained failure mechanism, it was shown that prognostics for the solder failure can be implemented by monitoring the resistance change in a thermal cycle condition.